Siglent Bench DMMs SDM3065X, SDM3055 and SDM3045X

[mawyatt]

Just did a quick check with our 2 custom references (blue cases) based upon LM399 (2) and LTZ1000, these references have well over a year 24/7 operation. DMMs have been ON for ~1/2 hr, and get similar relative readings with 10V output derived from both LM399s and LTZ1000, also get similar results when reading LM399 and LTZ1000 reference voltages. Get good results with scaled 1V and 0.1V from LM399.

On the 10V reference the SDM3065 reads consistently ~110uv high, on 1V is reads ~11uv high and on 0.1V scale  ~3.3uv low. Reading the reference voltage the SDM3065 reads ~50uv high.


Here's some images showing the readings. The far left is a AG34401A, the far right is a HP34401A, the KS34465A, SDM3065X and DMM6500 are in middle.

Best,

[mawyatt]

And 3 more measurements.

Best,

[TheDefpom]

about the calibrations, I did look at this site

https://www.thedefpom.com/siglent_sdm_calibration.php

thanks a lot for the link, he is doing the same, adding a user cal, you can select to use or not,
I called my work, got ok to pick up our brand new calibrated 34465A and the two siglent are now passed 24hrs
the keyside is on 3 hrs, plan is to let them all run, for a few days to see what happens, i take date spamped pictures a few times pr day..
I also got a friend with a HP3458A he said i could just pop in and pick it up, in case i like to try play with some more digits.

anyways the defpom site contain good explain videos about the calibration, and how to use, and select the factory or the user cal,
so i figured i try play with this, at least just for fun, just this one range, just to see if I can make them all read the same,
i might learn a few things along the way, or at least had some fun playing with my stuff.

anyways : i dont have any luck with this, my SDM3055 just say bad file name, to the CVS file from that site,
so how to fix this ? any one tried his csv files, on a 3055 model ?

by the way : thanks for the info about specifications, and expected error, i was aware of this
like the defpom he also up perform his unit, quite a lot with this type of user calibration

As I don't have the means to test on anything other than a 3065X the sites output files haven't been tested by me on other models, there could well be differences in how it handles the files.

The output files are named by the site for convenience, it could be that the firmware in the 3055 is handling this differently and expects exactly the right file names (my SDM3065X didn't care about the file name), so try renaming the file to "dc_voltage_recal.csv" and see if that helps, if it works (or not) let me know and I will adjust the site to suit.

[tautech]

@oz2cpu
If we need to Defpom can borrow a SDM3055 when our new stock arrives soon to check his Cal output files are correct.

[oz2cpu]

YES YES.. it finally worked.. it is VERY hard to make it work,
and the SDM3055 dont tell exactly what it does, it looks a little bit different,
at least it was able to read the file now, when renamed dc_voltage_recalA.csv when it is inside a folder of what ever name
there is a "thing" to do right, that is not clearly explained for beginners,
when i understand exactly what is the trick, i try explain it a bit more in detail here, ok ?

[oz2cpu]

ok here you go, see the two attached pictures, top left is the unit i play with

Manage File - Action much be RECALIBRATE !!
Browse - External Lower Dir - Find file
Select - Perform Recalibrate - YES - it will as if you like to use it, YES
DONE, back to voltmeter screen..

so now it reads correctly at 5V 10V 20V happy me.. but now.. it is wrong at zero input.. darn it,
the zero is not included in this file, it is something else ?
or will it be fixed when i do the negative part ?
i compleetly skipped the neg
UPDATE: zero, is simply the final result, of whatever you calibrate positive, and negative with, that means if both those
end points are done correctly, the zero, will be zero :-)
LOVE you all, stay cool..

[TheDefpom]

I checked the sites logs and could see that you only set a 5V cal point, that is too low, you need to set it higher than that, if you look at the recommended cal points on the site it says where the best points are to cal.

Yes you should definitely do both the positive and negative at the same time otherwise it will offset the entire range, make sure you set it back to factory cal before taking the measurements and doing the recalibration again.

[TheDefpom]

I will modify the site now to fix the file name issue you had so you shouldn't need to mess with it again.

[oz2cpu]

thanks a lot, fantastic site, and a great help
YES you are correct, do pos and neg cal right after each other, i did them a few times, by swapping banana blugs, and wrote down the best avarage I could find,
and bingo rock strady and correct readyout, my problem was only the 20 V range
all other ranges, all 3 meters track to the last digit,
while i can borrow the HP 34465A i go thu all other features the next days,
I leave them all on, no power downs.
during all this : i found a bug in the keysight, funny !! if you own one, try the DCV ratio, on and off, and see if the result go back ?? first time after power on ??
omg that sw dude must be the mongul of the year award :-)

[TheDefpom]

I think the file name issue could of been the length more than anything, I am not sure what the limitation is on file name length on the system the meters use, maybe it was just too long, or maybe it didn't like the "-" of the reference you used instead of the serial number?

Anyway as I don't want people to struggle with this in future I have set the site to use the default file names as used by Siglent, maybe they will enforce the proper file names in the firmware in the future, who knows, so I will play it safe and just use their naming system, it just means the meter model and serial number won't be part of the file name anymore.

[oz2cpu]

just tell me, if you like me to perform any type of file check experiments, i will be super happy to help out

[purpose]

Mr. Defpom... may you live for a thousand years.
I just had a pop at adjustment, forgot something and re did it.
I now have my 3055 pretty much identical to my 2015 for the first time in a few years.

Many thanks and Happy New Year.

Peter

[TheDefpom]

just tell me, if you like me to perform any type of file check experiments, i will be super happy to help out

if you could try using that default file name with _YOURSERIALNUMBER added to it IE: dc_voltage_recal_SDM30FAX2R1234.csv and seeing if it will accept that, if it does then it is probably safe to add that back in on the site.

[purpose]

BTW... What I like about this method is that you don't need pricey calibration equipment, only a meter you can trust and a stable power supply.
In my case a Keithley 2015 that hasn't been calibrated in 20 years and a DP832.
The higher ranges are out of reach, but just fire in what you read on your trusted gizmo as "actual calibrator value" and you're golden.
Have to slip back to the factory cal for the higher stuff, but that's what buttons are for.

purp

[Bad_Driver]

I‘m keen to see how your fresh recalibrated SDMs will behave during the next year!
 
To check my 3065 it would need access to a 7.5 digit DMM or a Fluke callibrator, no chance for me these days.
But I can look for a third 6.5 DMM and than start to wonder which one shows the right numbers 

The rabbit hole is in front of me…..

[purpose]

The rabbit hole is in front of me…..

You've already slipped in.

[Johnny B Good]

The rabbit hole is in front of me…..

You've already slipped in.

 That's really the truth of it.

 When you've invested some £700+ in a 6.5 digit bench meter simply to overcome the tempco issue of a couple of cheap 9999 counts Mestek DM91As you'd previously thought of as being very stable considering the 41 quid total invested, the best you can say is that you're now clinging onto the edge of that particular rabbit hole by your fingernails.

 I've had my 3065X permanently powered up ever since I took delivery of it a week into October and similarly the cheap AD584 voltage reference I'd bought about ten days later.

 The 3065X seems to suffer very little tempco since it tracked the slow downward decline of a μ7805 mounted on a temperature controlled baseplate (Rubidium oscillator reference project) without any indication of a tempco. The AD548, otoh, has only been consistent in that it's variation with temperature very closely tracks room temperature with no sign of any ageing related drift (at least over the last month of its 10 weeks run time to date).

 As has been previously mentioned, absolute accuracy is less important than consistency in the face of ambient temperature variations and ageing effects, particularly true with self contained hobbyist projects. Absolute accuracy is a bonus as far as hobby activities are concerned and only becomes important when comparing results between different lab setups in the wider world.

 The Chinese seller I'd bought the AD548 from was honest enough not to insult my intelligence by making any attempt at supplying a voltage calibration chart of any sort, not even the more likely accurate hand written type which I'd have still ignored anyway.

 At best, it could only be as good as the calibration of the 7 1/2 digit DMM used and the care taken to create such a chart - too many unknowns to invest much faith in afaiac. Far safer simply to assume that Siglent had been able to calibrate the 3065X to within its stated specifications and allow for the tolerances built into its specified calibration accuracy.

 After monitoring that very lightly loaded (20mA) temperature stabilised (36*C) μ7805 for nearly two months, I'm quite confident that the 3065X is well within its stated tempco specification. With just over 2000 hours run time clocked up (and some 1800 hours for the AD548), the worst of whatever ageing drift it had is now well and truly behind it.

 Absolute accuracy in a home lab is always going to have a question mark hanging over it but the stability seems to be as good as I could possibly have hoped for out of my SDM3065X investment. Stability trumps absolute accuracy every time simply because without it, trying to recalibrate to within a tight tolerance of absolute accuracy will be a waste of both time and money.

[Algoma]

Working with my SDM3055 today, Curious how the Trend Chart will show more details than what it will display numerically. While showing 0v on the display, it continues to chart at a resolution far beyond 5.5 digits.

 Well, given it's voltage reference, I'm likely looking at nonsense, but it remains curious how the ADC can resolve and chart it out anyway. Not Accurate or of any value, but curious that the ADC's resolution is willing to chart out nanovolts and picovolts onto the Trend chart when the input is close to 0V as possible.

[coromonadalix]

Maybe they should add a threshold level of somekind, to get rid of the noise floor ?? if its what the meter seems to measure ?

[Kleinstein]

The SDM3055 (and it looks like also the 3045) use an AD7190 ADC, that is capable of 24 bit resolution. So the numerical resolution of the ADC would be good for 7 digits.  With the SD ADCs the numerical resolution is the least important limit. The limiting factgors are more the linearity and the noise (here the AD7190 is quite good and well in the 6 digit range). The reference and linearity limit the usefull resoluion to some 5 to 6 digits.

Like many other modern meters the internal resolution is higher than what is displayed on the screen. Via the computer interfacer I would also expect to get those extra digits. They may still be helpfull, especially near zero, where the reference is not that relevant.

[oz2cpu]

here you go DEFPOM

renamed my test file to dc_voltage_recal_SDM30FAX2R1234.csv
and tried it on a fresh SDM3055, BINGO  accepted :-)
so i must have done something wrong,
you know the feeling ? all is easy, when you know what to do :-)

Anyways : is this also the feature request thread ? will it be monitored by Siglent ??
Problem : SDM3055 trend chart, time only works in sec, and 3600 sec is the max,
Feature suggest :
let the user select Sec, Min, Hrs with each values up to 3600
thanks in advance !!

[oz2cpu]

if you own one of the many supported multimeters, you will find the free java program
"Test Controller" very usefull.
it is explained more deeply in this thread : https://www.eevblog.com/forum/testgear/program-that-can-log-from-many-multimeters/
so no need to repeat it all again here, just go there and read about it..
attached screenshoot, i work with temperature controlled voltage referance experiments right now,
so the live, long time logging, own defined average filters and such, is VERY usefull for me.
note the SDM3055 i use, is only 5½ digits, but it send out all 7½ digits,
the last two looks like random, due to they did not low pass filter it..
but with this program i can add all the average i like, and actually get a few more digits stable.
see also the attached curve, this is filtered 7½ digits, 1 div = 1uV

[oz2cpu]

SIZE problems ?? manual not updated ?
pictures and measurement not correct ?
please see the attached pictures.
i dont have the 3065 anymore, but when i had one, (a loaner) i recall it was way deeper, than my 3055..
and it had exactly the same type of plastic corners.

my 3055 is 295mm from the end of the back foot, to the front plastic corner,
also my plastic rubber corners, are much bigger, than pictured.
the datasheet say 282mm

the 3065 is 293mm arcording to the datasheet, however it is very unclear from where is this is measured, see the picture,
i recall the 3065 dont fitt my table, and it also came in a bigger box.

so my question is : how deep is a new version of the SDM3065X with new type rubber back and front all included ?

[tautech]

so my question is : how deep is a new version of the SDM3065X with new type rubber back and front all included ?

SDM3065X has always sported the rubber bumpers.

From datasheet:
(length×width×height ): 345.45mm×260.29mm×107.21mm
https://int.siglent.com/u_file/document/SDM3065X_DataSheet_EN02G.pdf
Info on 3055 and 3045X here:
https://siglentna.com/download/18278/

Pic showing all 3 of which 3055 is an early version before the bumpers were upgraded to match the other models.
https://www.eevblog.com/forum/testgear/siglent-new-bench-dmm-sdm3055/msg1194407/#msg1194407



Also see comments in this post:
https://www.eevblog.com/forum/testgear/siglent-new-bench-dmm-sdm3055/msg1023105/#msg1023105

[oz2cpu]

thanks alot, this extra length is exactly as i remembered it,
this means the datasheet batronix link to, and uses for their SDM3065X is wrong :-)
the reason why i put this up, is one voltmeter is no enough for my experiments, so i am in the marked for one more,
and why not pick the other one, to be a lot better than the SDM3055 i allready got :-)
the timing issues reported on the SDM3065X is only relevant for people wanting to perform FFT or such timing demanding measurements,
it is as far as we hear only a sw thing, so i dont think i need to worry to much about it,
where i am now, i push my 3055 way beyond its specifications, and only this way i see minor drift on its own ref,
the 3065 got better ref, LM399 more stable, so i think it will solve the challenge i have for my experiments.
but it is too deep to fit my lab table .. oh them problems..